All schema retrieval code.

This service routes incoming query requests to query executor services with the required query language and database code running. It also allows for the retrieval of cached query results, which can be requested from the frontend. In the future we envision this service as a more advanced load balancer, as RabbitMQ only has round robin load balancing, which does not work if requests take arbitrary amounts of time.

The object store driver.

Here are the code and data files placed which I used for the publication in JGR:oceans about the effect of freshwater pulses on estuarine salinity.

Uncertainty-aware Personal Assistant for Making Personalized Privacy Decisions

A personal privacy assistant (PURE) helps its user make privacy decisions by recommending privacy labels (private or public) for given contents. PURE is unobtrusive, uncertainty-aware, and personalized.

In this repo, you can find how we implement PURE, Standard Neural Networks (SNN), and existing models such as Monte Carlo (MC) dropout and Deep Ensemble.

Code base of my Master's thesis.

This is a demo repository!

Extension of text_explainability for sensitivity testing (robustness & fairness).

A generic explainability architecture for explaining text machine learning models.

Generation base dependency

Project by Koray Poyaz and Irma Mastenbroek supervised by Michael Behrisch

BSc project by Casper Stiekema supervised by Michael Behrisch

(Very messy) codebase for my Master's thesis project on semi-supervised point cloud classification using label diffusion lidar segmentation